Z-Palatoplasty (ZPP): a Technique for Patients Without Tonsils

Z-palatoplasty (ZPP): A technique for patients without tonsils

MICHAEL FRIEDMAN, MD, HANI Z. IBRAHIM, MD, RAMAKRISHNAN VIDYASAGAR, MBBS, MS, JONATHAN POMERANZ, BS, and NINOS J. JOSEPH, BS, Chicago, Illinois

OBJECTIVE: Patients without tonsils and with Fried- In view of its limited success in curing OSAHS,1-3 man tongue position (FTP) III and IV are poor can- many adjunctive procedures have been proposed or didates for uvulopalatopharyngoplasty (UP3). Even performed concurrently or sequentially.3-5 The UP3 when combined with adjunctive hyopharyngeal technique was originally described by Fujito et al6 in techniques, results are poor. We assessed a modi- 1979, and, although many modifications have been fied uvulopalatoplasty based on a bilateral Z-plasty published, the basic procedure involves palate shorten- in treating patients without tonsils who have ob- ing with closure of the mucosal incisions, hence en- structive sleep apnea/hypopnea syndrome (OS- compassing “palatoplasty” component; classical tonsil- AHS). lectomy and pharyngeal closure comprise the METHODS: 25 patients treated with a modified tech- “pharyngoplasty” component of the procedure.7-9 nique were matched with 25 patients previously Several problems continue to exist: (1) No proce- treated with classic UP3. All patients in both groups dure has been studied for post-tonsillectomy patients. also had radiofrequency tongue base reduction. (2) Many patients, especially post-tonsillectomy pa- Preoperative vs. postoperative measures of objec- tients, end up with an extremely narrow palatal arch tive treatment success and subjective symptoms further contributing to airway obstruction. (3) Post- were compared for the 2 groups. Morbidity, includ- tonsillectomy patients have poor results with classical ing pain levels, narcotic use, and return to solid diet UP3.10,11 (4) A significant number of patients are not and normal activity, as well as complications were improved by UP3 but are actually made worse.2 studied. Patients who have had a previous tonsillectomy have RESULTS: Subjective improvement was good for an altered palatal anatomy that requires specialized both groups, but objective clinical improvement treatment. Often in these patients, the posterior tonsillar was significantly better for the experimental group. pillars have been resected or are scarred and the palate Morbidity and complications for the experimental is pulled closer to the posterior pharyngeal wall. In an group were comparable to the control group and effort to achieve maximal airway enlargement in 3 to other published series on UP3. areas, a new modified palatoplasty was developed. The CONCLUSIONS: A modified technique for patients goal was to widen the space between the palate and the without tonsils who have OSAHS is presented. The postpharyngeal wall, between the palate and tongue new technique is more successful with acceptable base and to maintain or widen the lateral dimensions of morbidity for patients with OSAHS than classical the pharynx. The new technique will be described and techniques. (Otolaryngol Head Neck Surg 2004; essentially represents a double Z-plasty to change the 131:89-100.) scar contracture tension line to an anterolateral vector and to widen the anteroposterior and lateral oropharyn- Uvulopalatopharyngoplasty (UP3) remains the most geal air spaces at the level of the palate, hence the term common surgical procedure performed as treatment for Z-palatoplasty (ZPP). obstructive sleep apnea/hypopnea syndrome (OSAHS). This retrospective/prospective study was designed to assess the safety and efficiency of this new procedure on 30 patients seeking surgical treatment for OSAHS From the Department of Otolaryngology and Bronchoesophagology, RushÐ and compare these results to a matched group of pa- Presbyterian-St. Luke’s Medical Center (Drs Friedman, Ibrahim, Vidyasa- gar, and Pomeranz), and Division of Otolaryngology, Advocate Illinois tients who previously underwent classical UP3. Of Masonic Medical Center (Drs Friedman, Ibrahim, Vidyasagar, and Joseph). these 30 patients, 25 completed the study with a min- Presented at the Annual Meeting of American Academy OtolaryngologyÐHead imum of 6 months of follow-up. and Neck Surgery, Orlando, FL, September 21-24, 2003. Reprint requests: Michael Friedman, MD, Department of Otolaryngology and Bronchoesophagology, RushÐPresbyterian-St. Luke’s Medical Center, 1653 MATERIALS AND METHODS West Congress Parkway, Chicago, IL 60612-3833; e-mail, njoseph@ Institutional review board approval for the study pentechassociates.com. protocol and appropriate informed consents were ob- 0194-5998/$30.00 Copyright © 2004 by the American Academy of OtolaryngologyÐHead and tained from 30 patients without tonsils and positive Neck Surgery Foundation, Inc. history, physical examination, and conclusive polysom- doi:10.1016/j.otohns.2004.02.051 nographic evidence of OSAHS.

Downloaded from oto.sagepub.com at RUSH UNIV on October 2, 2015 Otolaryngology– Head and Neck Surgery 90 FRIEDMAN et al July 2004

Thirty patients who were deemed surgical candi- Table 1. Friedman staging system based on dates and fitting the inclusion criteria were invited to Friedman tongue position, tonsil size, and body participate in the prospective (experimental) arm of mass index (BMI) the study and were scheduled to undergo a ZPP Friedman tongue Tonsil procedure. Prior to 2002, surgical treatment for pa- position size BMI tients with OSAHS consisted of UP3. A chart review Ͻ identified a matching set of 25 patients previously Stage I 1 3, 4 40 23,4Ͻ40 treated with standard UP3 with minimum 6-month follow-up. These 25 patients represented the retrospective (control) arm of the study. All the patients in both groups underwent adjunctive tongue base reduction by radiofrequency (TBRF) to address the hyopharyngeal narrowing. Stage II 1, 2 0, 1, 2 Ͻ40 3, 4 3, 4 Ͻ40 Inclusion Criteria Stage III 3 0, 1, 2 Ͻ40 All prospective patients who underwent surgical 4 0,1,2 Ͻ40 treatment of OSAHS had fulfilled previously reported criteria by the authors.10,11 In addition, selection criteria for the present study included: (1) no previous surgical treatment for OSAHS; (2) significant symptoms of snoring and/or daytime somnolence; (3) docu- Ͼ mented failure of continuous positive airway pressure Stage IV 1, 2, 3, 4 0, 1, 2, 3, 4 40 All patients with significant craniofacial or other trial; (4) documented failure of attempts at conservative anatomic deformities measures, such dental appliances when appropriate, change in sleeping position, and sleep hygiene; (5) From: Friedman M, Ibrahim I, Joseph NJ. Staging of obstructive sleep apnea/ hypopnea syndrome: A guide to appropriate treatment. Laryngoscope (In patient without tonsils or had underwent prior tonsil- Press). lectomy; (6) Friedman OSA stages II or III;10,11 (7) the appearance of obstruction at the level of the soft palate contributing to OSAHS (fiberoptic hypolaryngoscopy and Mu¬ller maneuver was performed on all patients); Physical Examination Parameters (8) proof of medical fitness adequate for surgery; and Patients underwent preoperative physical examina- (9) a clear understanding and expectations of the risks, tions included a full assessment of the upper airway morbidity, and likely outcomes of surgery. with nasopharyngolaryngoscopy, Mueller maneuver, and standard examination. In addition, patients were Preoperative Subjective and Quality-of-Life staged according to the previously described Friedman Evaluation staging system,11 based on the Friedman tongue posi- Candidates for surgical treatment of OSAHS were tion (FTP, formerly the Friedman palate position), ton- evaluated based on history. Patient histories included sil size, and body mass index (BMI) (Table 1). Weight assessments of snoring level (0-10) described by the and height were recorded at the initial visit and the BMI bed partner, Epworth Sleepiness Scale (0-24),12 and the (kg/m2) was calculated. Although originally named SF-36 v2 Quality-of-Life (QOL) score (0-100). The Friedman palate position (FPP), the term has been SF-36௡ v2 Health Survey (QualityMetric, Lincoln, RI) corrected and called Friedman tongue position (FTP) is a 36-item well-documented survey that has previ- because the observation that describes the palate/tongue ously been used to evaluate patients with obstructive relationship is, in fact, predictive of the tongue position sleep apnea. The survey consists of 8 multiitem health as it impacts the airway. domains: (1) physical functioning (PF); (2) role limita- Stage I disease was defined as those patients with tion as a result of physical health problems (RP); (3) FTP I or II, tonsil size 3 or 4, and BMI of less than 40 bodily pain (BP); (4) general health (GH); (5) vitality kg/m2 (Table 1). Stage II disease is defined as FTP I or (energy/fatigue) (VT); (6) social functioning (SF); (7) II and tonsil sizes 0, 1, or 2, or FTP III and IV with role limitation as a result of emotional problems (RE); tonsil sizes 3 or 4 and BMI of less than 40 kg/m2. Stage and (8) mental health (psychological distress and psy- III disease is defined as FTP III or IV, tonsil sizes 0, 1, chological well-being) (MH). A score of 0 to 100 is or 2, and BMI less than 40 kg/m2. All patients with a calculated for each domain based on patient responses. BMI greater than 40 kg/m2, regardless of FTP or tonsil A score of 100 represents the best possible health. size, as well as those patients with significant cranio-

Downloaded from oto.sagepub.com at RUSH UNIV on October 2, 2015 Otolaryngology– Head and Neck Surgery Volume 131 Number 1 FRIEDMAN et al 91

tion, the lateral dimension of the palate is usually doubled to approximately 4 cm. The exact dimensions of the flaps that extend in a butterfly pattern are illustrated in Figure 7. The anterior midline margin of the flap is halfway between the hard palate and the free edge of the soft palate. The distal margin is the free edge of the palate and uvula. The lateral extent is posterior to the midline and all the way the lateral extent of the palate. Prior to discharge, all patients were prescribed acet- aminophen with codeine elixir 12 mg/5 mL and directed to administer 15 cc q4 hr, as needed for pain. All patients also received postoperative antibiotics and steroids.

Adjunctive Surgical Procedures All patients also underwent TBRF (Somnoplastyª Fig 1. The incision of the palatal flap is marked. system, Gyrus, Inc., Memphis, TN). The patients received 3,000 joules distributed to 4 points along the midline of the tongue behind the circumvallate papillae. facial or other anatomic abnormalities were designated All patients also were asked to return to the office for as Stage IV (Table 1). monthly TBRF treatments. Each month, they received 1,500 joules distributed to 2 points along the midline. Polysomnography The total numbers of joules delivered are shown in An all-night attended, comprehensive sleep study Table 2. Patients were asked to return for treatments was performed using a computerized polygraph to until symptoms resolved and until their polysomno- monitor electroencephalogram (C3-A2, C4-A1), left gram indicated significant improvement of OSAHS. and right electrooculogram, electrocardiogram, chin Not all patients complied with this request. and anterior tibialis electromyogram, abdominal and thoracic movement by inductive plethysmograph, nasal Postoperative Follow-up oral airflow, oxygen saturation by pulse oximetry Postoperatively, patients were seen in the office at 1

(SaO2), and throat sonogram. Hypopnea is measured week, 2 weeks, at 1 month, monthly for follow-up via a thermistor placed in the path of airﬂow from the TBRF treatments, and at 6 months. At each interval nose and mouth. Apnea was deﬁned as cessation of examination and interview, the patients were queried breathing for at least 10 seconds. Hypopnea was a concerning any complications, other adverse effects, decreased effort to breathe at least 50% less than the and complaints. A postoperative polysomnogram was baseline and with at least a 4% decrease in SaO2. The scheduled at 6 months. At the 6-month follow-up or apnea-hypopnea index (AHI) was calculated as the sum later, patients were reassessed for snoring level (0-10) of total events (apneas and hypopneas) per hour. All as described by the bed partner, Epworth Sleepiness patients were studied in the same sleep laboratory fa- Scale (0-24), and the SF-36 v2 Quality-of-Life survey cility. (0-100).

Surgical Technique Statistical Analysis The surgical technique for the UP3 group has been All statistical analyses were performed using SPSS previously described by the senior author.13 The surgi- Version 11.0.1 (SPSS, Inc., Chicago, IL). Continuous data cal technique for the modified ZPP group is illustrated is displayed as means Ϯ standard deviation (SD). Statis- in Figures 1-8. The key points of the surgical technique tical significance was accepted when P Ͻ 0.05. The Stu- are as follows: (1) 2 adjacent flaps are outlined on the dent’s t- and the Mann-Whitney U-tests were employed to palate; (2) only mucosa of the anterior aspect of the 2 evaluate significant differences between ZPP and standard flaps is removed; (3) the 2 flaps are separated from each UP3 treated patients. The Levine’s Test for Equality of other by splitting the palatal segment down the midline; Variances was used to determine statistically significant (4) 2-layer closure bringing the midline all the way to variances. The paired Student’s t-test was used to compare the anterolateral margin of the palate is accomplished; preoperative vs. postoperative mean values within each and (5) the final result creates 3 to 4 cm of distance group. The chi-square test was used to test the association between the posterior pharynx and the palate. In addi- between categorical variables.

Downloaded from oto.sagepub.com at RUSH UNIV on October 2, 2015 Otolaryngology– Head and Neck Surgery 92 FRIEDMAN et al July 2004

Fig 2. The mucosa over the palatal ﬂap is removed, exposing the palatal musculature.

RESULTS Twenty-ﬁve patients ZPP patients were studied and matched with an additional 25 patients who had previously undergone UP3 for the treatment of OSAHS. Table 2 compares demographic data, including mean age, gender distribution, mean preoperative BMI, Friedman OSAHS stage as described by Friedman et al,11 OSAHS disease severity based on preoperative AHI (AHI Ͻ 15 ϭ mild, AHI 15-45 ϭ moderate, AHI Ͼ 45 ϭ severe), and treatment of tongue base via radiofrequency (total number of Joules delivered). There were no statistical differences between the 2 groups with regard to mean age, gender distribution, Friedman stage distribution, OSAHS severity, mean preoperative BMI, or TBRF joules delivered.

Adjunctive Treatment All patients in both groups received intraoperative and follow-up treatments to the tongue base with radiofrequency reduction. The total energy delivered was 4510 Ϯ 1874 joules in ZPP patients and 3840 Ϯ 1067.7 joules in UP3 patients. These values were not different from each other (Table 2).

Morbidity Fig 3. Lateral view of the soft palate and the uvula after the mucosa is excised. Note the uvula and palate are Postoperative. The number of days of narcotic pain hanging close to the posterior pharyngeal wall, narrow- medication usage and return to normal diet were used ing the retropharyngeal space. as indices of short-term morbidity from surgery. Pa- tients undergoing ZPP used narcotic pain medication for 6.4 Ϯ 3.6 days and required 6.4 Ϯ 1.9 days to return to a normal diet as compared with 9.4 Ϯ 2.7 days and 10.3 Ϯ 3.6 days, respectively in UP3 patients. These

Downloaded from oto.sagepub.com at RUSH UNIV on October 2, 2015 Otolaryngology– Head and Neck Surgery Volume 131 Number 1 FRIEDMAN et al 93

Fig 4. The uvula and palate are split in the midline with a cold knife.

Fig 5. The uvular flaps along with the soft palate are reflected back laterally over the soft palate. mean values for narcotic medication usage and return to between 2 and 60 days, and completely resolved by the normal diet were significantly different from each other fifth postoperative visit (3 months after surgery). All (Table 3). patients that complained of VPI noted occasional or Complications and Long-Term Morbidity. Peri- rare problem when drinking quickly. In no patient did operative complications were rare in both groups. the VPI significantly affect their ability to eat a normal Tongue base infections, secondary to TBRF treatments diet in social situations nor did it significantly affect that required antibiotic treatment were found in 1 ZPP their voice. No cases of permanent VPI were encoun- patient and 2 UP3 patients. None of these patients had tered in either group. The majority of complications en- airway obstruction or abscess formation that required countered were related to postoperative throat discomfort, surgical drainage. Temporary postoperative velopha- including globus sensation, mild dysphagia, dry throat, ryngeal insufficiency (VPI) was reported in the 12 ZPP and inability to clear the throat. Eleven ZPP and 17 UP3 and 7 UP3 patients. In all these patients, the VPI lasted patients reported such complaints (Table 3).

Downloaded from oto.sagepub.com at RUSH UNIV on October 2, 2015 Otolaryngology– Head and Neck Surgery 94 FRIEDMAN et al July 2004

Fig 6. Two-layered closure of the palatal ﬂaps. The submucosal layer is approximated ﬁrst with 2-0 Vicryl (polyglactin 910, Ethicon, Inc,, Somerville, NJ).

Fig 7. Two-layered closure of the palatal ﬂaps—the mucosal closure with 3-0 chromic suture.

Subjective Symptom Elimination subjective assessments of improvement in symptom Patients’ and bed-partners’ subjective assessment of severity between the 2 groups, the percent change in disease severity (snoring level and ESS) were collected snoring level and ESS was calculated using the follow- preoperatively and at the time of the 6-month postop- ing formula: erative follow-up examination. Only preoperative level ͑ Ϫ ͒ of snoring level differed, either preoperatively or post- Postop Preop ͫ% Change ϭ ͩ ͪ ϫ 100ͬ operatively, between the two groups (Table 3). How- Preop ever, postoperative values for both snoring level and ESS were signiﬁcantly lower than their respective pre- There were no differences in % change snoring level or operative values (Table 3). To more easily compare ESS between patients treated with ZPP or UP3 (Fig 9).

Downloaded from oto.sagepub.com at RUSH UNIV on October 2, 2015 Otolaryngology– Head and Neck Surgery Volume 131 Number 1 FRIEDMAN et al 95

of 5 or less. Table 4 compares the subjective improvement in OSAHS symptoms following surgical treatment of OSAHS with ZPP versus UP3. Subjective improvement based on snoring scale was encountered in 25 (100%) of ZPP patients and 24 (96%) of UP3 patients. There was no signiﬁcant difference between the 2 groups in subjective improvement of symptoms. Although mean postoperative ESSs were signiﬁcantly lower in both the experimental and control groups as compared with preoperative scores (Table 3), the percent changes in ESS for both groups were considerably less than the percent change in snoring level (Fig 9). Cross- tabulation analysis of the raw data showed that, particularly in the ZPP patients, a sizable number of patients (12 ZPP and 6 UP3) showed no improvement in their subjective ESSs following surgery.

SF-36 v2 Quality of Life Health Survey

Fig 8. Lateral view showing the widening of the nasophar- Twenty-five patients belonging to the experimental ynx after midline palatoplasty. group had completed the SF-36 v2 Health Survey both pre- and postoperatively, totaling 50 surveys. Scores from 0 to 100 (100 being the best health) were calculated for all Table 2. Demographic data 50 patients who 25 patients in each of the 8 domains both pre- and post- underwent either Z-palatoplasty (ZPP) or operatively. Figure 10 displays the preoperative vs. post- uvulopalatopharyngoplasty (UP3) for the operative mean scores (ϮSD) for each of the 8 domains treatment of obstructive sleep apnea/hypopnea for the experimental group. In addition, the difference in syndrome (OSAHS) mean score (ϮSE), postoperative vs. preoperative, for ZPP UP3 Sig. each of the eight health domains were calculated (Fig 11). -P-value) A positive mean difference in mean score represents im) (25 ؍ n) (25 ؍ n) provement, whereas a negative mean difference in mean Age (yrs) 49.7 Ϯ 12.6 50.4 Ϯ 9.6 NS Gender NS score represents deterioration in quality of life. Male 19 16 The postoperative improvement in mean scores was Female 6 9 statistically significant (P Ͻ 0.05) for 6 of the 8 do- Friedman OSAHS NS mains (Figs 10 and 11). Only PF (physical function) stage and BP (bodily pain) were not significantly improved. II 2 23 III 3 22 The greatest degrees of improvement were seen in the OSA severity NS RE (emotional role), MH (mental health), SF (social Mild 3 7 functioning), and VT (vitality/energy) domains with Moderate 11 10 mean percent increases equal to 13.0 Ϯ 20.1, 15.4 Ϯ Severe 10 8 17.4, 11.2 Ϯ 17.5, and 14.1 Ϯ 16.2, respectively. In BMI (kg/m2) 30.8 Ϯ 3.6 29.5 Ϯ 4.2 NS TBRF (total Joules 4510.0 Ϯ 1874.3 3840.0 Ϯ 1067.7 NS addition, positive mean differences were also seen in delivered) the RP and GH domains (Fig 11).

*Staging system for grading OSAHS as previously described by Friedman et Objective Surgical Success al.8 OSA severity based on AHI (AHI Ͻ 15 ϭ mild; AHI 15Ð45 ϭ mild; AHI Ͼ 45 ϭ severe). BMI, body mass index; TBRF, base of tongue reduction by Objective measure of clinical improvement of OS- radiofrequency. Statistical significance accepted when P Ͻ 0.05. NS, not AHS was based on data collected during polysomnog- significant. raphy. Specific indicators included the apnea index (AI), the apnea/hypopnea index (AHI), and the mini-

mum recorded arterial oxygen saturation (Min SaO2). Because nearly all the patients originally sought Table 5 compares mean (ϮSD) preoperative vs. post- treatment for loud snoring, we also determined subjec- operative values for the ZPP patients and UP3 patients. tive improvement of the patients’ symptoms using a In both ZPP and UP3 patients, mean AI and AHI values strict criteria, which required a 50% decrease in snoring decreased and mean Min SaO2 increased postopera- level postoperatively and a postoperative snoring level tively as compared with their preoperative values. Ex-

Downloaded from oto.sagepub.com at RUSH UNIV on October 2, 2015 Otolaryngology– Head and Neck Surgery 96 FRIEDMAN et al July 2004

Table 3. Comparison of subjective indices of disease severity (snoring level and Epworth sleepiness scale) and postoperative course (narcotic medication days, return to normal diet, and morbidity) in patients undergoing Z-palatoplasty (ZPP) and uvulopalatopharyngoplasty (UP3)

ZPP UP3 Sig. (P-value) (25 ؍ n) (25 ؍ n)

Snoring level (1Ð10) Preoperative 9.6 Ϯ 0.6 7.4 Ϯ 1.0 P Ͻ0.001 Postoperative 2.6 Ϯ 2.1* 2.4 Ϯ 2.3* NS Epworth sleepiness scale (1Ð24) Preoperative 12.5 Ϯ 6.2 14.2 Ϯ 3.6 NS Postoperative 8.3 Ϯ 4.0* 8.7 Ϯ 4.3* NS Narcotic pain meds use (days) Postoperative 6.4 Ϯ 3.6 9.4 Ϯ 2.7 P Ͻ0.005 Return to normal diet (days) Postoperative 6.4 Ϯ 1.9 10.3 Ϯ 3.6 P Ͻ0.002 Tongue base infection Postoperative 1 (4%) 2 (8%) NS Bleeding Postoperative 0 (0%) 0 (0%) NS Postnasal drip Postoperative 3 (12%) 4 (16%) NS Dysphagia Postoperative 1 (4%) 11 (44%) P Ͻ0.001 Foreign body sensation Postoperative 11 (44%) 17 (68%) NS Temporary velopharyngeal insufﬁciency Postoperative 12 (48%) 7 (28%) NS Permanent velopharyngeal insufﬁciency Postoperative 0 (0%) 0 (0%) NS

Statistical signiﬁcance accepted when P Ͻ 0.05. *Statistical signiﬁcance from the preoperative value.

Table 4. Comparison of successful surgical treatment and improvement of symptoms of OSAHS between Z-palatoplasty (ZPP) and uvulopalatopharyngoplasty (UP3) in selected patients

ZPP UP3 Sig. (P-value) (25 ؍ n) (25 ؍ n)

Objective success 17 (68.0%) 7 (28.0%) P ϭ 0.005 Subjective improvement 25 (100%) 24 (96%) NS

Statistical signiﬁcance accepted when P Ͻ 0.05.

cept for preoperative AI, other preoperative or all postoperative polysomnogram data did not differ between the 2 groups (Table 5). Using the classic deﬁnition of successful surgical Fig 9. Comparison of Z-palatoplasty (ZPP) vs. uvulopalatopharyngoplasty (UP3) for the treatment of obstructive treatment of OSAHS, which requires a 50% or greater sleep apnea/hypopnea syndrome on % change of post- reduction in postoperative AHI as compared with the operative vs. preoperative values in snoring level and preoperative value and a postoperative AHI of less than Epworth Sleepiness Scale (ESS). Both snoring level and ESS 20, we determined the success or failure of ZPP or UP3 decreased signiﬁcantly postoperatively in both groups, in each patient (Table 4). Surgical treatment of OSAHS however, the % change in each group were not different from each other. with ZPP combined with TBRF resulted in successful treatment in 17 (68%) as compared with 7 (28%) success in patients treated with UP3 combined with TBRF.

Downloaded from oto.sagepub.com at RUSH UNIV on October 2, 2015 Otolaryngology– Head and Neck Surgery Volume 131 Number 1 FRIEDMAN et al 97

Fig 11. Percent change of postoperative vs. preoperative quality-of-life values for each of the 8 SF-36 v2 health domains in 25 patients treated with Z-palatoplasty (ZPP) for the treatment of obstructive sleep apnea/hypopnea Fig 10. Comparison postoperative vs. preoperative qual- syndrome. A positive mean difference in mean score ity-of-life values for each of the 8 SF-36 v2 health domains represents improvement, whereas a negative mean dif- in 25 patients treated with Z-palatoplasty (ZPP) for the ference in mean score represents deterioration in quality treatment of obstructive sleep apnea/hypopnea syn- of life. Statistical significance accepted when P Ͻ .05. drome. Statistical significance accepted when P Ͻ .05. *Denotes significant difference between mean preoper- *Denotes significant difference between preoperative ative and postoperative scores. PF ϭ physical function, RP and postoperative values. PF ϭ physical function, RP ϭ ϭ physical role, BP ϭ bodily pain, GH ϭ general health, VT physical role, BP ϭ bodily pain, GH ϭ general health, VT ϭ vitality/energy, SF ϭ social functioning, RE ϭ emotional ϭ vitality/energy, SF ϭ social functioning, RE ϭ emotional role, and MH ϭ mental health. role, and MH ϭ mental health. A score of 0-100 is calculated for each domain based on patient responses. A score of 100 represents the best possible health. Table 5. Comparison of polysomnographic data before and after Z-palatoplasty (ZPP) or uvulopalatopharyngoplasty (UP3) in 50 patients These values were significantly different from each treated for sleep apnea/hypopnea syndrome other (Table 4). ZPP UP3 Sig. (P-value) (25 ؍ n) (25 ؍ n) DISCUSSION The value of UP3 as an isolated procedure for treat- BMI (kg/m2) ment of OSAHS has been questioned by many studies Preoperative 31.0 Ϯ 3.1 29.6 Ϯ 4.2 NS Ϯ Ϯ 1-3,5,11 Postoperative 31.4 3.5 29.8 4.5 NS because of variable results. Its role as part of a AI comprehensive treatment plan that includes adjunctive Preoperative 14.8 Ϯ 18.6 4.6 Ϯ 8.8 P Ͻ0.017 procedures, however, remains solidly accepted in most Postoperative 3.5 Ϯ 8.2* 2.8 Ϯ 5.4 NS situations in which the palate is contributing to airway AHI turbulence and obstruction. The goal of this study was Preoperative 41.8 Ϯ 26.4 33.4 Ϯ 13.9 NS Postoperative 20.9 Ϯ 19.3* 25.2 Ϯ 16.6* NS to focus only on the palatal component, and keep ad- Min SaO2 (%) junctive treatment “standard” in both the experimental Preoperative 81.3 Ϯ 10.6 84.4 Ϯ 8.9 NS and control groups. We treated tongue base with radio- Postoperative 87.8 Ϯ 8.4* 86.6 Ϯ 7.4 NS frequency reduction, but any other treatment of the BMI, body mass index; AI, apnea index; AHI, apnea/hypopnea index; Min hypopharynx could have been used as a “standard” for SaO2, minimum arterial oxygen saturation. both groups. Our hypothesis was that by designing a Statistical significance accepted when P Ͻ 0.05. better palatoplasty we could improve subjective and *Statistical significance from the preoperative value. objective results without increasing morbidity. The goal of UP3 is to widen the airspace in 3 areas: (1) the retropalatal space, (2) the space between tongue are the most difficult in whom to achieve a “squared base and palate, and (3) the lateral dimensions. The off” wide palatal area with the standard technique. The results however often fall short of this goal in all 3 nature of standard UP3 brings the posterior palatal areas. Patients who have had a previous tonsillectomy mucosal forward, narrowing the oropharyngeal inlet at

Downloaded from oto.sagepub.com at RUSH UNIV on October 2, 2015 Otolaryngology– Head and Neck Surgery 98 FRIEDMAN et al July 2004

the tip of the uvula is amputated to remove the excess mucosa and expose muscle. The mucosa over the uvula and some of the soft palate are removed with a scalpel. Finally, the uvula is reflected back toward the soft palate and sutured. This procedure was designed to treat snoring while minimizing the possible risk for VPI. It was not specifically designed as a treatment for OSAHS. The concept, however, of preservation of the palate and using it as a flap was the basis of our modified technique. A thorough review of the international literature identified an article by Mauro B. M. Vieira et al7 entitled Zetapalatoplasty. The technique describes a form of Z-plasty that requires intact anterior and posterior pillars and is meant for patients with intact tonsils. In this modified technique, tonsillectomy is first performed, followed by a lateral posterior incision made on the soft palate from the superior point of the Fig 12. Traditional uvulopalatopharyngoplasty. The direc- tonsil site to create a triangular flap. A second incision tion of pull is anteromedial, eventually narrowing the is made on the middle part of posterior tonsillar pillar retropharyngeal airway in the midline. directed upward and medially, creating a superomedial- based triangular flap. The tip of the superomedial-based flap is then pulled and sutured on the soft palate inci- the level of the new free edge of the palate with a sion. The same procedure is repeated on the other side. resulting triangular shape of the palate rather than the Finally, subtotal uvulectomy is done. The procedure originally desired “squared shape.” As further contrac- was designed mainly to prevent VPI following classical tion occurs, additional narrowing further adversely af- UP3. Because the tonsillar pillars are often partially fects long-term results (Fig 12). The Z-palatoplasty sacrificed in previous tonsillectomies, this technique technique changes the direction of contracture from was not suitable for our group of patients. This article superior-medial to superior-lateral (Fig 12). was identified after completion of our series. The prin- Our experience shows that patients without tonsils ciple of changing the direction scar contracture by having OSAHS are found to be poor candidates for UP3. Z-plasty is similar to our concept. Our procedure, how- It may be because of the damage caused to the posterior ever, is far more aggressive and is designed for maxi- pillar by previous tonsillectomies, which resulted in scar- mal improvement. ring and thereby pulling the soft palate toward the poste- Fairbanks,14 in his extensive study of UP3 reported rior pharyngeal wall and thus, often do not have redundant that the palatal incompetence was attributed to exces- pharyngeal folds. The “pharyngoplasty” component of sive resection of the palate, particularly in the midline. UP3 is designed to increase pharyngeal space by first Palatal closure depends on the central mounding action removing hypertrophied tonsils and secondly by eliminat- of the musculus uvulae and on the lifting action of the ing redundant pharyngeal folds. Including a pharyngeal levator palati muscles, which course from the eusta- component for these patients usually only adds morbidity chian tube downward (posteriorly) and medially to without any benefit. Most surgeons already limit surgery enter diagonally in the midline. Hence, our procedure on these patients to a uvulopalatoplasty and eliminate the was modified in such a way that all muscles are pre- “pharyngeal” component. Fairbanks14 has contributed sig- served and distortion of midline tissue is minimized. nificantly to improved results by recommending that the In our modified technique, the midline of the soft posterior tonsillar pillar be advanced lateral cephalad. palate is retracted anterolaterally, which results in a When patients have had previous tonsillectomy with re- widened retropalatal area. The uvula is split in midline section or scarring of the posterior tonsillar pillar, this and sutured laterally along with the adjacent soft palate, important step is not possible. thereby creating an effective anterolateral pull on the Various modifications of the UP3 have been previ- soft palate and thus widening the retropalatal area. ously proposed.7-9 The reversible uvulopalatal flap de- Additionally, because the muscles of soft palate are scribed by Powell et al8 and its modification, the uvu- preserved, risks of complications such as permanent lopalatal flap,8 involve reflecting back the uvula back VPI are minimized. This is crucial because the tech- toward the softÐhard palate junction. First, a portion of nique results in a more anterior position of the palate

Downloaded from oto.sagepub.com at RUSH UNIV on October 2, 2015 Otolaryngology– Head and Neck Surgery Volume 131 Number 1 FRIEDMAN et al 99

(Fig 9). However, when objective success was exam- ined, our modified ZPP technique showed considerable improvement over classical UP3. Objective success was based on improvement in postoperative versus preoperative polysomnograms. Seventeen of 25 (68%) of ZPP patients were classified as clinically successful based on a minimum 50% reduction in AHI and a postoperative AHI of less than 20. The UP3 group had a success rate of 28% that is consistent with previous studies (Table 4).5 The modified ZPP technique also resulted in significantly less postoperative morbidity as compared to UP3. Modified ZPP patients used pain medications for fewer days (6.4 Ϯ 3.6) and required fewer days to return to normal diet (6.4 Ϯ 1.9) than those patients treated with UP3 (9.4 Ϯ 2.7 and 10.3 Ϯ 3.6 days, respectively). This may be explained by the fact that the Fig 13. Z-palatoplasty. Note the anterolateral direction of palatal mucosa was only partially excised, sparing the pull on the soft palate that widens the retropharyngeal musculature and no pharyngeal muscle was exposed. space. The options available for evaluating quality of life in patients with OSAHS include disease-specific tools, such as the Calgary Sleep Apnea Quality of Life Index and VPI is a major concern. Our modified technique (SAQLI), and generic tools, such as the SF-36 v2. results in a dramatically improved postoperative ap- Lacasse et al15 demonstrated that although the SAQLI pearance of the pharynx. The anteroposterior and lat- has strong content and construct validity and is more eral space is significantly larger than comparable areas responsive to changes in quality of life than the SF-36 with the classic UP3. The line of healing and contrac- v2, it must be administered by an interviewer, is time- ture is anterolateral (an oblique vector with some ver- consuming, and demonstrates redundancy. We have tical but predominantly horizontal components), so that found the SF-36 v2 Health Survey to be a reliable long-term healing and contracture will continue to survey that is shorter, generic self-completed question- widen the airway rather than narrow it, as can occur naire with well-documented validity and has been pre- with the classical UP3 (Figs 12 and 13). viously used in patients with OSAHS.16-18 We measured the results of surgical treatment by The SF-36 v2 is designed to evaluate patients’ quality categorizing patients according to subjective and objec- of life in 8 domains of health. Using a scale of 0 to 100, tive improvement. The subjective success was based on patients with high scores in a particular domain have a comparative improvement on snoring level, Epworth better quality of life in that domain. The SF-36 v2 is Sleepiness Scale, and quality-of-life questionnaires, designed so that raw scores can be used in isolation and/or pre- and postoperatively in each group. Previously, we be compared to national norms.19 In our study, we used have shown that UP3 with adjunctive TBRF has excel- the raw scores alone, because each patient was compared lent subjective results.5 Thus, it is not surprising that pre- and postoperatively against him or herself. Using the ZPP treatment with adjunctive TBRF could not im- raw data to compute scores in each of the 8 domains, we prove on these results. Our results were excellent for were able to compare quality of life preoperatively and improving snoring and qualty-of-life data. Many of our postoperatively (in each of the 8 domains) within the patients, however, did not improve when ESS was used experimental group. Because the control group was taken as a guide. We suspect that many of these patients were from the charts retrospectively, we were unable to com- seeking relief from snoring, and were unaware of day- pare between the 2 groups time somnolence. Their preoperative ESS scores were The experimental group showed a significant degree therefore relatively low. The assessment (ESS) is based of improvement in the postoperative quality of life. on subjective information and is subject to the patient’s Specifically, patients reported greater improvement in awareness of his or her daytime function and his or her VT (vitality/energy), which correlates with the im- willingness to identify problems. If the preoperative provement in snoring and OSAHS. The improvement score is low, it is not surprising that a postoperative in SF (social functioning) correlates with their ability to score will show no improvement. This is reflected in sleep well with their bed partners. Both these factors relatively low percent changes in ESS in both groups indirectly led to an improvement on the emotional and

Downloaded from oto.sagepub.com at RUSH UNIV on October 2, 2015 Otolaryngology– Head and Neck Surgery 100 FRIEDMAN et al July 2004

psychological well-being as shown by the improvement jectively) in treating the palatal level obstruction in OS- in RE (role emotional) and MH (psychological distress AHS patients without tonsils. We suggest that Z-palato- and psychological well-being). plasty might serve as a potential alternative to the The modified technique however suffers from many traditional UP3 in treating the palatal level obstruction in limitations. It is a more aggressive change of the palate, patients without tonsils. As with any new procedure, a despite the fact that only mucosa is removed. Preser- learning curve will lead to improved results. Additional vation of the palatal musculature probably contributes studies are needed to examine the efficacy of Z-palato- to the absence of permanent VPI in our patients. It does plasty in the treatment of patients with tonsils. not mean, however, that the procedure is reversible if VPI should occur. The more aggressive treatment REFERENCES clearly resulted in a higher incidence of temporary VPI 1. Sher AE, Schectman KB, Piccirillo JF. The efficacy of surgical in our study group. modifications of the upper airway in adults with obstructive sleep A very significant limitation is the absence of any apnea syndrome. Sleep 1996;19:156-77. 2. Senior BA, Rosenthal L, Lumley A, et al. Efficacy of uvulopala- clear landmark to describe the size of the flaps. The topharyngoplasty in unselected patients with mild obstructive extent of mucosal removal at the midline is half way sleep apnea. Otolaryngol Head Neck Surg 2000;123:179-82. between the hard palate and the limit or resection in a 3. Riley R, Guillominault C, Powell N, et al. Palatopharyngoplasty classical UP3. The lateral component of the flap ex- failure, cephalometric roentgenograms, and obstructive sleep apnea. Otolaryngol Head Neck Surg 1985;93:240-4. tends further back and all the way laterally as illus- 4. Powell NB, Riley RW, Guilleminault C. Radiofrequency tongue trated. These are guidelines and the technique clearly base reduction in sleep-disordered breathing: a pilot study. Oto- requires operator judgment. The procedure is techni- laryngology 1999;120:683-6. cally more difficult and takes longer than the classic 5. Friedman M, Ibrahim H, Lee G, et al. Combined uvulopalatopharyngoplasty and radiofrequency tongue base reduction for UP3. This technique, like the classical one, results in treatment of obstructive sleep apnea/hypopnea syndrome. Oto- the absence of a uvula and many patients complain of laryngol Head Neck Surg 2003;129:611-21. a “foreign body” sensation in the throat. 6. Fujita S, Conway W, Zorick F, et al. Surgical correction of anatomic abnormalities in obstructive sleep apnea syndrome: This study also suffers from limitations. It is a rel- uvulopalatopharyngoplasty. Otolaryngol Head Neck Surg 1981; atively small group with short follow-up. It is ex- 89:923-34. tremely difficult to obtain postoperative polysomno- 7. Viera MBM, Leite SHP, Cunha FC. Zetapalatopharyngoplasty grams on successfully treated patients. The longer the (ZPPP): new surgical technique for snoring and sleep apnea treatment. Braz J Otolaryngol 2001;67:1-8. time interval from surgery, the more difficult that tasks 8. Huntley T. The uvolopalatal flap. Op Tech Otolaryngol Head becomes. Therefore, the 6-month cutoff was used as a Neck Surg 2000;11:30-5. compromise to be able to capture a reasonable sample 9. Powell NB, Riley RW, Guilleminault C, et al. A reversible of patients. After 6 months, usually only the patients uvulopalatal flap for snoring and sleep apnea syndrome. Sleep 1996;19:593-9. whose treatment failed are willing to undergo addi- 10. Friedman M, Tanyari H, La Rosa M, et al. Clinical predictors of tional testing. Although some of the patients may ulti- obstructive sleep apnea. Laryngoscope 1999;109:1901-7. mately relapse, the study was designed to compare two 11. Friedman M, Ibrahim H, Bass L. Clinical staging for sleep- procedures both at 6 months. It clearly shows the ben- disordered breathing. Otolaryngol Head Neck Surg 2002;127:13- 27. efit of the modified technique. 12. Johns MW. Daytime sleepiness, snoring, and obstructive sleep Finally, an ideal study would have been entirely apnea. The Epworth Sleepiness Scale. Chest 1993;103:30-6. prospective and randomized. After the initial pilot stud- 13. Friedman M, Landsberg R, Tanyeri H. Submucosal unvulopala- ies with the technique, however, it became obvious that topharyngoplasty. Op Tech Otolaryngol Head Neck Surg 2000; 11:26-9. it was far superior to the classical technique. It was 14. Fairbanks DN. Operative techniques of uvulopalatopharyngo- therefore felt it was wrong to design a study that would plasty. Ear Nose Throat J 1999;78:846-50. not allow half of our patients to benefit from the im- 15. Lacasse Y, Godbout C, Series F. Independent validation of the proved technique. sleep apnoea quality of life index. Thorax 2002;57:483-8. 16. Flemons WW, Reimer MA. Development of a disease-specific CONCLUSIONS quality of life questionnaire for sleep apnea. Am J Resp Crit Care Med 1998;158:494-503. Obviously, no single procedure would be effective in 17. Jenkinson C, Stradling J, Peterson S. Comparison of three mea- treating all OSAHS patients. Treatment should to be tai- sures of quality of life outcome in the evaluation of continuous positive airways pressure therapy for sleep apnea. J Sleep Res lored according to the anatomy of each patient, as dem- 1997;6:199-204. onstrated in the present study. We conclude that rerouting 18. Friedman M, Ibrahim HZ, Ramakrishnan V, et al. Modified the uvula and soft palate laterally can more effectively uvulopalatoplasty: a modified technique in selected patients. enlarge the retropharyngeal space and improve the airway Laryngoscope 2004;114:441-9. 19. Ware JE, Kosinski M, Dewey JE. How to score version 2.0 of the characteristics as compared with the traditional UP3. Our SF-36 Health Survey. Lincoln, RI: QualityMetric Incorporated; modification appears effective (both subjectively and ob- 2000.

Downloaded from oto.sagepub.com at RUSH UNIV on October 2, 2015